Lambda the Ultimate

Ehud missed a golden opportunity to plug OpenOffice; it reads this Word file and also handles PowerPoint and Excel. Since Ehud mentioned his disdain for the "Microsoft-centric nature of C#" see my plug for non-Microsoft C# goodies. Not that I have a beef against Microsoft, but they should have called it C-sharp-elbow in deference to their business practices. If you've never heard of Microsoft's "kill the baby" strategy, or studied their financials, take the time to get acquainted.

looks far less ugly
Not as clean as Nice (should that be as nice as Clean?)

Seems like more features of C# are intended to allow the compiler to catch potential errors: out parameters, non-virtual methods by default, explicit overide and new method keywords, checked integer arithmetic (yipee!) ...

LtU regulars know I like Liskov and Guttag's Program Development in Java. Will they write a C# edition?

Ehud: Please notice that my question is about MS ownership of C# as such, but rather about how it's going to fare in the education market.

Yes, but MS ownership may have much to do with the answer to that question. Read Wired, ZDNet.

Note that I admire the .NET runtime architecture, and respect Microsoft Research, if not the whole corporation. C#'s impact on education will be the usual function of Microsoft hegemony. Hasn't it been your experience that schools tend to teach what industry asks them to teach? Or is your attention on specific technical differences that will affect how students come to understand what OO is all about?

My prediction re the success of C# stems from the fact that I assume students will demand to be taught the "industry standard" language.

I assume you mean this discussion. OK.

students will demand to be taught the "industry standard" language
And if that's what they want there are now plenty of vocational courses.

If students are choosing between CS programmes according to which PLs are used on courses, then that seems to be a statement about how poorly they understand or value the bulk of the programme.

Using that criterio a curriculum which didn't cover either Java or C# would be doing students a disservice. Assuming that is that those students want to get jobs in corporate software development.

Short list of non-obvious things determined from the C# 2.0 draft

- Overloading on return values is rather tricky. The question is how you go about convincing the compiler which type is the proper return type. The author sites Eiffel, but I vaguely recall that Eiffel does not have overloading at all (though I could be misremembering). IIRC, Meyer spurned function overloading on types as being too confusing.

- Nullable types is rather difficult if you want to maintain primitive types within the language. C# does have boxing/unboxing but it does introduce some inconsistencies along the way.

- Type inferencing is just in the nature of generics. If you are going to use generics, then it's only proper that you know how the compiler translates into the instance types.

- Partial types are intended more for the code generators than they are for human consumption. MS wants to seperate the environment maintained methods/properties and the programmer supplied ones. The intention is not aimed for compiled partials. As such, I don't see why the complaint.

Well, I can't say that C# will overtake Java in the schools, but I can say without doubt that they ought not.

Besides, the language that you learn to code in is very unimportant. That ought to be the easy part about studying computer programming. These languages they are all the same.

The only distinction I really make is between the functional and the imperative programming languages. That's why I read LtU in the first place!

But seeing as the languages are rather similar, I don't see a reason to say that without doubt Java should not be replaced by C#.

What's the big deal?

The decisive battle is between VMs, methinks. On that note a Java VM for .NET.